Technical method estimate of the numbers of micro-organisms present in the original sample. Pathogenic organisms isolated from patients producing mucopurulent sputa were frequently present in virtually pure culture along the 107 per ml dilution streak (fig 3). However, 'coliforms' cultured from sputa produced in patients who were already receiving broad-spectrum antimicrobial therapy were sometimes isolated at 108 per ml. Examination of the Gram film and clinical -assessnent of the patient were necessary before deciding upon their significance. One case of Escherichia coli pneumonia was seen in this series, and the colony count in both dilution and loop streak methods was greater than 109 per ml. We acknowledge the technical assistance of Mr. A. Porter and Miss E. M. Jones from this laboratory,

and Mr. S. P. Willmott at the Public Health Laboratory, Epsom, Surrey. References Lindsey, D. (1959). Quantitative bacteriologic study of tissues, fluids, and exudates. J. Lab. clin. Med., 53, 299-307. Miles, A. A. and Misra, S. A. (1938). The estimation of the bactericidal power of the blood. J. Hyg. (Camb.), 38, 732-749. Monroe, P. W., Muchmore, H. G., Felton, F. G., and Pirtle, J. K. (1969). Quantitation of microorganisms in sputum. Appl. Microbiol., 18, 214-220. Pirtle, J. K., Monroe, P. W., Smalley, T. K., Mohr, J. A., and Rhoades, E. R. (1969). Diagnosis and therapeutic advantages of serial quantitative cultures of fresh sputum in acute bacterial pneumonia. Amer. Rev. resp. Dis., 100, 831-838. Sharpe, A. N. and Jackson, A. K. (1972). Stomaching: a new concept in bacteriological sample preparation. Appl. Microbiol., 24, 175-178.

Letters to the Editor Isolation of Vibrio cholerae

agglutination in a slide test, there is seldom any difficulty in identifying the vibrio within one day with reasonable I would like to add two suggestions to the confidence. lucid instructions by Furniss and H. K. GHOSH Donovan (J. clin. Path., 1974, 27, 764Microbiology Department, 766) for the isolation and identification Royal Newcastle Hospital, of the cholera vibrio. Australia 2300 One is that if the thiosulphate-citratebile-salt-agar (TCBS) medium is preferred to Monsur's, one should also use a non-inhibitory medium, because we SI units, pH and cH+ find that the dehydrated TCBS from one We were disappointed in the UK-SI British supplier is fairly inhibitory even Committee's reply (Baron et al, 1975) to fresh isolates and becomes almost to the comments by Bold and his colcompletely inhibitory about two weeks leagues (1975) regretting the lack of after the container has been opened. guidance as to whether an SI unit should Storage may be better in a cooler climate, be used to report H+ concentration. but this needs to be defined. A simple It is precisely because it is an area of non-selective medium can be made from scientific controversy that expert guidance peptone agar or nutrient agar with 0-5% would have been appreciated. This parbile salt or 0-1% Teepol adjusted to pH ticular topic has after all been contro8-08-5. versial ever since Sorensen (1909a,b) deThe second point is the importance of fined pH formally ascolony morphology. There are very few 1 faecal bacteria having the flat, almost log [H+] transparent, bluish colonies of the cholera vibrio. This is of course better seen on a whereas in actual practice transparent medium than on TCBS. A rare strain may have a wrinkled centre pH = pH. + (E32-3026 RT on overnight incubation: a rugose variant. However, the rest of the colony The debate as to whether the special will give a clue. With a positive oxidase symbol pcaH should be used to denote reaction and the usually strongly positive H+ activity as opposed to H+ concentra-

RF

tion continued until Peters and van Slyke (1931), in their classic work, pronounced in favour of the commonsense use of pH to cover both situations. Our own preference for reporting H+ in acid-base work in nanomole/litre has been expressed elsewhere (Howorth, 1974), but a further reason for resolving the arguments about activity/concentration of ions is the increasing use of electrodes other than the H+-glass electrode in chemical pathology. For example, we expect to have in routine use within a few months a non-colorimetric serum electrolyte analyser based entirely on specific ion electrodes: Na+ electrode : for Na+ : for K+ K+ $, H+ 9,, : for total CO2 ,, : forurea NH3 : for glucose ,, 02 Many other specific ion electrodes, for example, for calcium and halides, are routinely used elsewhere, as well as CO2 and 02 electrodes in acid-base work. If the assumptions about activity coefficients implicit in converting electromotive activity to concentration units are unacceptable to the UK-SI Committee and others, we would all have to consider reporting Na+ results as pNa, K+ as pK, and so on (Svendsen, 1966). We ask, is it really a useful concept to report urea as

424

Book reviews

'pNH3' when using a potentially very Book reviews accurate method: urea

-,-

NF13 + C02

ureaseV

'-specific ion electrode for NH3 but report urea as millimole/litre when using a non-specific measurement of serum diacetyl monoxime reactants? One might also ask en passant whether any of the UK-SI Committee report serum osmolality as millimole/kilogram in their laboratories using instruments which actually measure depression of freezing point. The relevant point is surely that all our ordinary techniques have errors of one sort or another. We are sure our clinical colleagues want their results expressed in a uniform easilv understandable manner (one of the few virtues of the SI) and certainly do not want different units used according to the physical principles underlying a particular technique (colorimetry/flame photometry/specific ion electrodes in the present case). P. J. N. HOWORTH AND A. D. HIRST

Department of Chemical Pathology, King's College Hospital Medical School, Denmark Hill, London SE5 8RX References Baron, D. N., Broughton, P. M. G., Cohen, M., Lansky, T. S., Lewis, S. M., and Shinton, N. K. (1975). Problems in the introduction of SI units (Letter). J. clin. Path., 28, 164-165. Bold, A., Hoffenberg, R., London, D. R., Whitehead, T. P., and Wilding, P. (1975). SI units. (Letter). J. clin. Path., 28, 164. Howorth, P. J. N. (1974). RIpH revisited. Lancet, 1, 253-254. Peters, J. P. and Van Slyke, D. D. (1931). Quantitative Clinical Chemistry, Vol. 1, Interpretations, p. 872. Balliere, Tindall and Cox, London. Sorensen, S. P. L. (1909a). Enzymstudien. It. Ueber die Messung und die Bedeutung der Wasserstoffionenkonzentration bei enzymatischen Prozessen. Bioc-hem. Z., 21, 131-200 and 201-304. Sorensen, S. P. L. (1909b). Erganzung zu der Abhandlung: Enzymstudien. 11. Biochem. Z., 22, 352-356. Svendsen, A. (1966). Simplification of electrolyte analyses by means of Na and K electrodes. Proc. Ass. clin. Biochem., 5, 66-68.

A Colour Atlas of Microbiology (Wolfe Medical Atlases, II) By R. J. Olds. (Pp. 288; 398 illustrations. £7-00.) London: Wolfe Medical Books. 1974. It is not entirely clear for whom this book is intended. The author says that it is primarily for students of medicine and paramedical sciences, but, as the text is very brief, a considerable background knowledge is required to understand it, and the atlas must be used in conjunction with a textbook of methods. It consists of 279 colour photographs which illustrate the end products of various identification procedures. Some sections, particularly those on fungi and biochemical tests, would be useful to an experienced laboratory worker performing an unfamiliar test or suspecting an unfamiliar organism, but the photographs of bacterial colonies do not always show the differential characteristics as intended. This is a pity, since a useful function of the atlas would be as an alternative to cultures of dangerous pathogens or rare organisms for teaching purposes. The text is purposely brief, but in some instances needs a little expansion. For example, a photograph of mycobacteria

stained by the Ziehl-Nielsen method 'illustrates why treatment with strong acid or alkali is used to decontaminate sputum' without a comment that the stain is selective for acid-alkali-fast bacteria. In spite of these reservations it is a pleasant book to browse through and find some techniques which are not performed routinely in one's own laboratory and, although it has limited value as a teaching manual, it is worth having as a reference book for occasional use. B. JAMESON

Hemorrhagic Diseases and the Pathology of Hemostasis. By Armand J. Quick (P. xiii + 380; illustrated. $23.75.) Springfield, Illinois: Charles C. Thomas. 1974.

The

This distinguished man, one of the pioneers in blood coagulation, was working before 1935, and now in his 80th year he still has the wide view of the physiologist and clinician and an appreciation of our old masters-his favourite seems to be William Osler. This is a work for the

expert and the connoisseur and not for the student or physician seeking a modern balanced view of the subject. This book is as much a distillation and a defensive justification of one man's work as a review of the haemorrhagic diseases and haemostasis. Almost inevitably he presents these great themes seen through the eyes of one man looking back over a long active life in which history will probably decide that the biggest ideas came early. Over and over again he seeks to elucidate the problems almost exclusively with the use of the prothrombin time, the prothrombin consumption test, the bleeding time, and the aspirin tolerance test. Well presented with a great wealth of clinical detail all culled from this one man's experience, this is a fine valedictory tome by the originator of the Quick prothrombin time. J. R. O'BRIEN

Anaerobic Bacteria: Role in Disease Edited by Albert Balows, Raymond M. DeHaan, V. R. Dowell, Jr, and Lucien B. Guze. (Pp. xxi + 655; illustrated. $27.50.) Springfield, Illinois: Charles C. Thomas. 1974.

This book is a collection of papers given at an international symposium held in Atlanta, USA, in November 1972. It has four editors and 66 contributors of whom six were British, five from other European countries, and one Canadian. There is much here of great interest on the experimental side, particularly the recent work on the role of intestinal anaerobes in malabsorption syndromes and colonic carcinoma; also the importance of adherence to mucous surfaces in the mouth and experimental evidence of increased pathogenicity of mixtures of anaerobes. In contrast the clinical side is a little disappointing. It was stated repeatedly that anaerobes will not be isolated without 'good technique' but just what this is was not made clear and no recommendation of how to test one's technique emerged. Indeed, it was implied that a high isolation rate was the best indicator but, of course, this depends very much on the clinical condition of the patients from whom the specimens are received which varies considerably between different hospitals and in different countries. It was stressed by at least two contributors that specimens in danger of contamination are not suitable for excamina-